Immigrant Children in High and Low LEP Schools · 1 Throughout this research, LEP will include both students who are limited English proficient (LEP) and students who do not know

Who’s Left Behind? Immigrant Children in

High and Low LEP Schools

Clemencia Cosentino de Cohen

Nicole Deterding

Beatriz Chu Clewell

Program for Evaluation and Equity Research The Urban Institute

Washington, DC

September 2005

THE URBAN INSTITUTE

The Urban Institute is a nonprofit, nonpartisan policy research and educational organization established in Washington, D.C., in 1968. Its staff investigates the social, economic, and governance problems confronting the nation and evaluates the public and private means to alleviate them. The Institute disseminates its research findings through publications, its web site, the media, seminars, and forums. Through work that ranges from broad conceptual studies to administrative and technical assistance, Institute researchers contribute to the stock of knowledge available to guide decisionmaking in the public interest. Conclusions or opinions expressed in this report are those of the authors and do not necessarily reflect the views of officers or trustees of the Institute, advisory groups, or any organizations that provide financial support to the Institute.

Program for Evaluation and Equity Research (PEER)

The Urban Institute’s Program for Evaluation and Equity Research (PEER) focuses on education research. PEER staff conduct studies in the fields of educational attainment, educational access, minorities in mathematics and science, teacher education, teacher recruitment and retention, and educational assessments. Much of PEER’s work has centered around evaluation studies—many of these large, multisite, multi-method evaluations of programs to increase educational access and success among underrepresented groups, as well as programs to increase the teaching pool. Support for PEER comes from multiple sources, including the National Science Foundation and the American Association for the Advancement of Science, as well as private foundations, such as the Ford Foundation, DeWitt Wallace Reader’s Digest Fund, Lumina Foundation, and GE Foundation.

Copyright © 2005. The Urban Institute. All rights reserved. Except for short quotes, no part of this report may be reproduced or used in any form or by any means, electronic or mechanical, including photocopying, recording, or by information storage or retrieval system, without written permission from the Urban Institute.

____________________________________________________________________________________________________

THE URBAN INSTITUTE Program for Evaluation and Equity Research (PEER) Page iv

C O N T E N T S

About this Research …………………………………………………………………………………… v

Acknowledgments …………………………………………………………………………………… v

EXECUTIVE SUMMARY …………………………………………………………………………………… 1

INTRODUCTION …………………………………………………………………………………… 2

METHODOLOGY …………………………………………………………………………………… 3

FINDINGS …………………………………………………………………………………… 4

1. Schools: How Do High-LEP Schools Differ from Low- and No-LEP Schools? 4

2. Principals: How Do Principals in High-LEP Schools Differ from Principals in Low- and No-LEP Schools? 8

3. Teachers: How Do Teachers in High-LEP Schools Differ from Teachers in Low- and No-LEP Schools? 10

CONCLUSIONS …………………………………………………………………………………… 14

DISCUSSION …………………………………………………………………………………… 15

IMPLICATIONS …………………………………………………………………………………… 16

References …………………………………………………………………………………… 18

Appendices

Appendix A: List of Figures and Tables

Appendix B: Statistical Tables

______________________________________________________________________________________

THE URBAN INSTITUTE Program for Evaluation and Equity Research (PEER) Page v

ABOUT THIS RESEARCH

In an effort to expand knowledge about young immigrant populations and to examine how the No Child Left Behind Act (NCLB) affects schools serving limited English proficient (LEP) students, the Urban Institute was funded by the Foundation for Child Development to complete a series of reports.

The series includes four reports:

• A demographic profile of LEP children and children of immigrants with a special focus on pre-K to 5th grade, an extension of the Urban Institute’s previous demographic analysis of immigrant children under the age of 6, also supported by the Foundation for Child Development.

• A road map document decoding the complexity of NCLB and its impacts on LEP students, with a focus on elementary schools.

• A report of standards implementation in High-LEP elementary schools, based on case studies in three school districts, that aims to shed light on how NCLB has influenced strategies for meeting the needs of LEP students and the children of immigrants.

• A statistical portrait comparing elementary schools with high concentrations of LEP students to schools with fewer or no LEP students to examine differences that may affect schools’ abilities to meet NCLB requirements.

The last report—the profile of elementary schools with varying concentrations of LEP students—is presented herein. As is true of the other components of this research series, this report was informed by the knowledge gained over the course of the entire project.

ACKNOWLEDGMENTS

This report was made possible by generous financial support from the Foundation for Child Development.

We gratefully acknowledge insightful comments from Tamara Lucas, Michael Fix, Randy Capps, and the Foundation for Child Development. We also thank Rob Santos for his technical advice regarding the statistical analysis.

______________________________________________________________________________________________

THE URBAN INSTITUTE Program for Evaluation and Equity Research (PEER) Page 1

EXECUTIVE SUMMARY

The No Child Left Behind Act of 2002 (NCLB) is a federal law that, among other things, holds schools accountable for the academic performance of limited English proficient (LEP) students. Many of these LEP children are immigrants or children of immigrants. The law’s stringent requirements, which converge with rapid demographics-driven change in the U.S. immigrant population, necessitate renewed attention to how immigrant children and the children of immigrants are educated in the United States.

Using data collected in the 1999-2000 Schools and Staffing Survey (SASS), this report studies the characteristics of schools serving immigrant children at the time of NCLB's passage. As SASS lacks a measure of immigration status among school children, this analysis uses English language proficiency level (or LEP status) as a proxy for immigrant status, recognizing that some LEP students may be neither immigrants nor their first-generation children, but rather second- or third-generation U.S.-born children. Focusing on nationally-representative information about elementary schools, principals, and teachers, the study compares differences between schools with high proportions of LEP students and schools with fewer and no such students to examine potential differences among the schools educating the nation's young.

The analysis reveals that LEP elementary school students are largely concentrated in a small number of schools: nearly 70 percent of the nation’s LEP students are enrolled in 10 percent of its schools. These schools, called in this report “High-LEP,” are predominately located in urban areas, and LEP students are largely minority and economically disadvantaged, embedding the discussion of LEP students’ education within the context of what is already known about poor, minority, immigrant-serving urban schools. The results also show that almost one-third of all LEP children enroll in schools serving low percentages of LEP children, or “Low-LEP” schools. Important differences between High-LEP and Low-LEP schools are observed in principal and teacher demographics, training, and experience. Teachers in High-LEP schools are more likely than those in other schools to have provisional, emergency, or temporary certification, and new teachers in High-LEP schools are substantially more likely to be uncertified. On the other hand, Low-LEP schools lag behind High-LEP schools in LEP-focused in-service training for general education teachers, and in offering important services (such as support and enrichment programs). From these and other findings, a complicated picture emerges. Concerns regarding the resources and services available to immigrant children are in some instances allayed, and in others reinforced, not only for students in High-LEP schools but also for the nearly one-third of LEP students in Low-LEP schools. The report concludes by discussing the implications of the findings, focusing in particular on the educational opportunities of LEP students in the context of the No Child Left Behind Act.

______________________________________________________________________ THE URBAN INSTITUTE Program for Evaluation and Equity Research (PEER) Page 2

INTRODUCTION

In January 2002, George W. Bush signed the No Child Left Behind (NCLB) Act into law. This legislation holds schools accountable for the performance of all students, including limited English proficient students (LEP) or English language learners (ELLs).1 LEP students tend to be immigrants or the children of immigrants (second- and in some instances third-generation).2 Our research focuses on these children. In the absence of data measuring the immigrant status of children in U.S. schools, and recognizing that the terms are not perfectly interchangeable, LEP status is used in this analysis as a reasonable proxy for immigrant status.

Today, limited English proficient students are the most rapidly growing population in U.S. schools. In the ten years between 1993 and 2003, the share of English language learners at elementary and secondary schools increased by over 50 percent, from 2.8 to more than 4 million children.3 Many states experienced even higher growth rates of 200 percent or more (Alabama, Arkansas, Colorado, Indiana, North Carolina, and others). This should not be surprising, given record-high immigration rates over the same period and a shift away from traditional receiving states (such as Florida) in favor of other destinations (such as Arkansas). In light of these trends and the resulting changing demographics of U.S. elementary schools, this research takes a closer look at the educational opportunities offered to English language learners. We wish to discern whether ELL students tend to be concentrated in certain types of schools, as previous research suggests (Ruiz-de-Velasco and Fix 2001), and whether the schools ELL students attend are as well-equipped as other schools to comply with the new requirements imposed by NCLB.

These two critical issues—namely, LEP concentration and adequacy of school resources—are addressed in this analysis by categorizing schools according to the percentage of LEP students they enroll. We then use this classification to compare schools with varying concentrations of ELL students. Using data from the 1999–2000 Schools and Staffing Survey (SASS)—a nationally representative survey of schools—public elementary schools were grouped into three categories: High-LEP, Low-LEP, and No-LEP.

The analysis presented in this report, under Findings, compares average school, principal, and teacher characteristics across these three types of elementary schools to indicate potential differences in the provision of educational services. Details regarding the methodology and data set used are provided in the next section, Methodology. A summary of findings is presented in the Conclusions section, while an analysis of findings is found under Discussion. The final section, Implications, presents a discussion of implications of findings for the education of children who are limited English proficient in the context of NCLB.

1 Throughout this research, LEP will include both students who are limited English proficient (LEP) and students who do not know English at all (NEP). The terms ELL and LEP will be used interchangeably. 2 Urban Institute research found that about 35 percent of LEP students in pre-K to 12th grade are first-generation immigrants, 46 percent are second-generation (children of immigrants), and 19 percent are third- or more generations (this category includes Puerto Ricans) (Capps et al. 2005). 3 Figures based on the 50 states and District of Columbia (Padolsky 2005).


METHODOLOGY

Schools and Staffing Survey (1999–2000): Public Elementary Schools

The data used in this report come from the 1999–2000 Schools and Staffing Survey (SASS) sponsored by the National Center for Education Statistics (NCES). SASS is the most comprehensive survey of elementary and secondary schools, school administrators, and teachers available. In 1999–2000, data were collected from approximately 4,700 school districts, 12,000 schools (traditional public, public charter, private, and Bureau of Indian Affairs schools), 12,300 principals, 52,400 teachers, and 9,900 library media centers. The “traditional public school” data used in the present analyses are representative, once appropriate weights are applied, of all public (non-charter) schools in the United States. Because this work focuses on public elementary schools, analyses were restricted to public schools generally serving students in pre-K to grade 5 (for additional information, see Defining Elementary Schools, below). Once weighted, the resulting data set represented 49,826 public elementary schools, serving nearly 23 million children, of whom about 2 million are limited English proficient.

Advantages and Disadvantages of the SASS Data Set

The Schools and Staffing Survey—the only data set containing the information necessary for this study—has the advantage of allowing us to produce nationally representative statistical estimates. The latest available data, used in the present analyses, were collected in the 1999–2000 school year, a year and a half before NCLB’s passage. While this time lapse may appear a disadvantage in terms of understanding current capacity issues—and how these issues might interact with NCLB provisions—it provides a useful pre-NCLB baseline measure of the conditions likely present in schools when NCLB was passed. Schools are institutions that do not change rapidly, and it is therefore unlikely that the conditions present in 2000 were much different than those present when NCLB was enacted in 2002.

Defining Elementary Schools

This study centered on elementary schools serving students in pre-K to 5th grade. To use the SASS data and ensure nationally representative statistical estimates, we had to adopt the definition of “elementary” schools created by NCES. This definition encompasses the grades of interest to this research, but it includes some schools that also serve grades 6 to 8. As a result, some schools in our analysis include students who are not in pre-K to 5th grade. To determine how closely the NCES definition of elementary schools matches the target group of our study, we calculated the share of each school’s students that was pre-K to 5th grade. Assuming equal numbers of students at each grade level (SASS does not provide grade-level enrollment information), approximately 92 percent of students served in NCES-defined elementary schools are in the grades of interest to our study.

Defining Schools by LEP Enrollment

Elementary schools were categorized according to the percentage of their students identified as limited English proficient. The resulting typology included three main categories: High-LEP, Low-LEP, and No-LEP schools (see table 1). “High-LEP” schools were defined as the top 10 percent of elementary schools in terms of LEP enrollment, representing the approximately 5,000 elementary schools where LEP students account for about 25 percent or more of all students enrolled.4 “Low-LEP” schools were defined as those where LEP students represent less than a quarter of all students, and include close to 24,000 schools (47 percent of all elementary schools). “No-LEP”

4 The actual cutoff was 23.5%. See table 1 for more information.


schools are those that report enrolling no students who are limited English proficient, and add up to about 21,000 schools (43 percent of all elementary schools).

Table 1: Elementary Schools by LEP Enrollment

Elementary Schools Total Enrollment LEP Enrollment N % N % N %

All 49,826 100 22,637,370 100 2,049,678 100 LEP Category

No-LEP 21,277 42.7 7,596,360 33.6 0 0 Low-LEP 23,566 47.3 12,040,539 53.2 643,492 31.4

High-LEP 4,983 10.0 3,000,471 13.3 1,406,186 68.6 Source: Urban Institute analysis of NCES School and Staffing Survey, 1999–2000. Notes: No-LEP = no LEP children enrolled at these schools; Low -LEP = < 23.48% children are LEP; High-LEP = =23.48% of children are LEP.

Analyzing Factors of Interest by LEP Enrollment

The above typology constitutes the main methodological tool used to create this profile of LEP-serving schools. Each factor of interest—such as teacher certification, principal educational level, and programs offered—is studied by producing average statistics for schools with no LEP children, schools with low percentages of LEP children, and schools with high percentages of LEP children. We thus report information about each school type and compare across the three LEP categories to see if statistically significant differences are found between High-LEP schools and those with fewer or no LEP students. Given the complex sampling design of SASS, Taylor Series Linearization was used to estimate sampling errors and test for statistical significance.5

FINDINGS

This section presents the analyses of responses from schools, principals, and teachers to the SASS survey questions. Data tables referenced in this section are included in the appendices. Unless otherwise noted, all differences reported here are statistically significant. Actual significance levels are included in the data tables.

1. How Do High LEP Schools Differ from Low- and No-LEP Schools?

The majority of LEP students are concentrated in a small number of schools: Nearly 70 percent of LEP students nationally enroll in only 10 percent of elementary schools. In these schools, LEP students account for almost one half of the student body (on average), a striking contrast to the 5 percent of LEP students enrolled in the average Low-LEP school (tables 1 and 2, figure 1). As described earlier, High-LEP schools are those where LEP students account for 23.5 percent or more of the student body. By this definition, High-LEP schools represent the top 10 percent of schools in terms of concentration of LEP students, about 5,000 schools. Alternately, Low-LEP schools have less than 23.5 percent of their student body classified as LEP, and No-LEP schools do not enroll any children who are limited English proficient. The data therefore show that LEP students are highly concentrated at a few schools. At High-LEP schools, LEP 5 Taylor Series Linearization is a statistical procedure used to correct bias in standard errors and significance levels introduced when data are obtained through complex sampling designs rather than simple random sampling.


students account for about 47 percent of the student body, while at Low-LEP schools they account for 5 percent. This suggests that the majority of ELL students are segregated in schools serving primarily ELL and immigrant children. As further testament to this condition, nearly half of elementary schools in the United States have no limited English proficient students.

Figure 1. Concentration of LEP Students in Elementary Schools

Source: Urban Institute analysis of NCES Schools and Staffing Survey, 1999-2000.

43% schools

0% students

47%schools

High LEP Low LEP No LEP

10% schools

31% students

69% students

High-LEP schools tend to be large and urban, while Low- and No-LEP tend to be smaller and predominantly suburban and rural, respectively (table 2, figure 2). Enrollment at High-LEP schools tends to be higher than enrollment at Low-LEP and particularly No-LEP schools (about 600 versus 500 and 350 students, respectively). This difference is partly explained by location. High-LEP schools are more likely to be in urban areas than the other school types; about 46 percent of High-LEP schools are in urban settings, versus about 24 percent of the Low-and No-LEP schools. Low-LEP schools, on the other hand, are more apt to be in suburban areas, while No-LEP schools are more often found in rural parts of the country (41 percent of No-LEP versus 19 percent of Low-LEP and 10 percent of High-LEP schools are in rural areas).

High-LEP and No-LEP schools have almost perfectly identical, but inverse, distributions of student enrollment by race: students in High-LEP schools are mostly minority, while students in Low- and No-LEP schools are predominantly White (table 2, figure 2). At High-LEP schools, minority students account for 77 percent of the student body, and White students for the remainder. The opposite is true at No-LEP schools, where White students make up 76 percent of the student body. Low-LEP schools are somewhat more evenly distributed, enrolling 35 percent minority students. Not surprisingly, these differences are largely driven by Hispanic students, who make up 53 percent of the students enrolled in High-LEP schools. As expected, the other racial minority more likely found in High-LEP schools is Asians, who comprise about 10 percent of enrollment in High-LEP schools versus 4 and 1 percent, respectively, of enrollment in Low- and No-LEP schools. Black students are the only minority group more likely to be in Low- and No-LEP than in High-LEP schools.


Figure 2. Demographic Description of Elementary SchoolsLEP concentration, predominant location, poverty, and student ethnicity

Free and Reduced Lunch Eligible

Asian Hispanic Native American White


Predominant>45% Urban >50% Suburban >40% Rural

LEPHigh LEP Schools Low LEP Schools No LEP Schools

3.5%

Concentration:

Poverty:

1.6%

75.7%

Black

65.2%

18.9%

3.3%

23.0%

0.9%

Race/Ethnicity:

Location:

9.9% 10.9%

15.6%14.1%

3.9% 1.3%

72%

40% 41%

52.8%

The incidence of poverty and health problems is significantly higher in High-LEP than in other schools (tables 2 and 4, figure 2). The percentage of poor children, as represented by the share qualifying for free and/or reduced-price school lunches, is significantly higher in High-LEP schools (72 percent) than in either Low- or No-LEP schools (about 40 percent). The high concentration of Hispanic children in urban areas—many of them immigrants or children of immigrants—helps explain the high incidence of poverty in High-LEP schools. Urban Institute researchers estimate that of the 11 million immigrant children and children of immigrants accounted for the by 2000 Census, about half were low-income (Capps et al. 2005). Confirming the significance of this finding, poverty is cited as a “serious problem” by more than 40 percent of principals and teachers at High-LEP schools, versus 20 percent or less of staff at other schools. Student health problems, likely also related to poverty, are identified as “serious” and “moderate” more frequently in High- than in Low- or No-LEP schools. More than 30 percent of principals and 45 percent of teachers in High-LEP schools ranked student health problems as “serious” or “moderate,” versus about 17 percent of principals and 21 percent of teachers in Low- and No-LEP schools.

High-LEP schools are more likely to offer Title I services. Regardless of LEP concentration, Title I services are more often provided schoolwide than as targeted assistance (table 2). More than 80 percent of High-LEP schools offer Title I services, versus 60 to 70 percent of Low- and No-LEP schools. This indicates that High-LEP schools are indeed more likely to have funding available to provide needed support or supplemental services. Within each of the three types of schools, Title I services are more likely to be offered schoolwide than as targeted assistance, particularly in High-LEP schools. The ability to offer services schoolwide extends the benefits of Title I–funded services


to all students, and simplifies reporting requirements for the schools. High-LEP schools, however, are no more likely than other types of schools to offer targeted assistance Title I services. Low-LEP schools, prevalent in suburban (and therefore likely wealthier) areas, are the most likely not to offer Title I services (40 percent of Low-LEP schools versus 30 percent of No-LEP and 18 percent of High-LEP schools do not offer Title I services).

Instructional contexts vary significantly by LEP school type: While High-LEP schools have slightly larger class sizes, these schools are more likely to offer support and remedial programs (pre -K, enrichment, after-school, summer school) (tables 2 and 3, figure 3). Not surprisingly, given the finding that High-LEP schools are larger than other types of schools, the average class size at High-LEP schools is two to three students higher than in Low- and No-LEP schools. While not benefiting from smaller class sizes, students in High-LEP schools do enjoy more special programs. These schools are more likely to have a pre-K program on site (43 percent of the High-LEP schools have such a program versus 30 to 33 percent of schools in the other LEP types). They are also significantly more likely to offer academic support programs—including summer school programs, whether for remedial or advancement purposes, and before- and after-school academic enrichment programs. In addition, High-LEP schools have a higher incidence of foreign language immersion programs, likely building on the language abilities ELL students bring with them to the classroom.

Figure 3. Special Programs Offered

5%

High Low No High Low No High Low No High Low No High Low No


43%

65%

Schools by LEP Concentration

30%33%

59%

73%

12%

27%22%25%

34%

49%

83%

74%

Foreign Language Immersion

Remedial Programs (intersessions)

Pre-K Program on Site

Academic Enrichment (before/after school)

Enrichment Programs (intersessions)

Native language instruction is more prevalent in High- than Low-LEP schools. The difference in use of other LEP-targeted instructional techniques, though significant, is less marked (table 3). Unsurprisingly, the data suggest, in High-LEP schools, instruction is more likely to be adapted to the needs of ELL students than is true in other schools. Techniques used to impart language and subject matter instruction include structured immersion, bilingual education, English as a second language (ESL), and native language instruction. The largest differences between Low- and High-LEP schools are in use of native language instruction, whether to maintain the language or to teach different subjects. About 43 percent of High-LEP schools rely on native language instruction to maintain students’ native tongues or to teach them subject areas, a practice common in less than 15 percent of Low-LEP schools. Differences across schools in other forms of instruction—ESL, bilingual—are less marked, with about 82 percent of Low-LEP and 94 percent of High-LEP schools using these instructional techniques.


Standardized procedures for identifying ELLs are more prevalent in High- than Low-LEP schools (table 3). A greater proportion of schools in the High-LEP versus the Low-LEP category report using external, standardized means of identifying English language learners, such as achievement tests, language proficiency tests, and home language surveys. Among Low-LEP schools, teacher observations or teacher referrals are more frequently used to identify students as English language learners than in High-LEP schools.

High-LEP schools are more likely to be involved in parental outreach and support activities than schools with lower concentrations of LEP students (tables 4 and 5). Eighty-five percent of High-LEP schools versus 65 percent of Low-LEP schools report engaging in parental outreach. Strategies include assigning a staff as parent liaison (prevalent in 76 percent of High-LEP, 60 percent of Low-LEP, and 55 percent of No-LEP schools), providing child care and/or transportation to facilitate parent participation in school activities or events (66 percent of High- versus 51 percent of Low- and 41 percent of No-LEP schools), and having a parent drop-in center (36 percent of High- versus 28 percent of Low- and 23 percent of No-LEP schools). In High-LEP schools, teachers are also more frequently required to involve parents in schooling, which they accomplish by sending letters explaining lessons or suggesting parental activities, and giving homework assignments that require parental participation. These activities—and the fact that they are more likely to be offered in High-LEP schools—may partly be a response to Title I regulations, which require substantive parental involvement for schools receiving Title I funds. But they may also be a response to a perceived problem with parental involvement and student behavior. Teachers, and to a lesser extent principals, at High-LEP schools more frequently report that parental involvement is a moderate to serious problem than their counterparts at Low- and No-LEP schools. Staff at High-LEP schools are also more likely to indicate that tardiness, absenteeism, and lack of preparation among students are moderate or serious problems, again perhaps partly explaining increased school attempts to involve parents.

High-LEP schools face more difficulties filling teaching vacancies and are more likely to rely on unqualified and substitute teachers than schools with few or no LEP children (table 5). Schools with high concentrations of LEP students are more likely to encounter difficulties filling teaching vacancies than schools with no or low concentrations of LEP children. This finding is suggested by the fact that High-LEP schools are more likely to hire unqualified teachers (37 percent of these schools do, versus about 11 percent of other schools) and are almost twice as likely to rely on substitutes than schools with fewer or no LEP students.6 Asked to report how difficult filling teaching vacancies was, 12 percent of High-LEP schools versus 5 percent of Low- and No-LEP schools indicated “very difficult.” In addition, 47 percent of High-LEP schools and almost 70 percent of Low- and No-LEP schools indicated positions were “easy to fill.”

2. How Do Principals in High-LEP Schools Differ from Principals in Low- and No-LEP Schools?

A growing body of research indicates that principals are a key explanatory factor of school success (Waters et al. 2003; Bryk et al. 1998; Clewell and Campbell 2004; Hallinger and Heck 1996, 1998; Sebring and Bryk 2000). For this reason, this study explores principal characteristics at each type of school to see if differences are found in such important factors as education and training.

6 Information drawn from the following SASS survey question: Did this school use the following methods to cover [teaching] vacancies? (1) Hired a fully qualified teacher, (2) Hired a less-than-fully qualified teacher.…(8) Used long-term or short-term substitutes. In this research, category 1 is considered qualified, category 2 unqualified, and category 8 substitute.


Principal Demographics and Compensation

Principals in High-LEP schools are more racially diverse than the national average. Principals in schools serving LEP children are also more likely to be female than are those in No-LEP schools (table 6). Principals in High-LEP schools are substantially less likely to be White (58 versus over 80 percent), more than five times more likely to be Hispanic, and more likely to be Asian and American Indian than are principals in schools with fewer or no LEP students. About the same proportions of principals in all three types of schools are Black. Looking at the proportion of principals of a given racial or ethnic background nationwide, about 42 percent of all Hispanic principals and 55 percent of all Asian principals work in the 10 percent of schools that are High-LEP. Also, a higher proportion of schools with LEP children have female principals (63 percent in High-LEP and 57 percent in Low-LEP schools) than schools with no limited English proficient children (49 percent).

Principals in schools with high concentrations of LEP students earn more than their counterparts in schools with fewer or no LEP students.7 This finding is true even once school location (urban, suburban, rural) and principal experience and education are taken into account. The average salary premium for principals at High-LEP schools is about $2,500 when compared to the earnings of principals at Low-LEP schools and over $6,500 when compared to principals in schools with no LEP children.

Principal Education and Experience

Principals at Low- and No-LEP schools have higher credentials on average than principals at High-LEP schools, although the latter are more likely to hold a PhD (table 6). The majority of principals at all three types of schools hold mater’s degrees. At High LEP schools, however, thirteen percent of principals indicated that their highest degree was a doctorate, compared with 8 to 10 percent of principals at other schools. On the other hand, principals in Low-and No-LEP schools are more likely than those in High-LEP schools to have completed post-MA education specialist degrees or professional diplomas. Hence, combining all types of post-MA degrees, about 36 percent of principals at High-LEP schools have higher-level training, versus 45 percent of principals at Low-and No-LEP schools.

Principals in High-LEP schools have fewer years of experience as principals (on average), but roughly the same amount of prior teaching experience, than principals in schools with fewer or no LEP students (table 6). On average, principals in High-LEP schools have almost a year less experience at their current school than Low-LEP principals, and almost two years less than principals in schools with no LEP students (4.25 years versus 5 and 6 years, respectively). When total principal experience is considered—that is, at their current and at prior schools—High-LEP principals have roughly two years less experience than do principals in No-LEP schools, but their overall experience approximates that of principals in Low-LEP schools. Combining teaching experience prior to and since becoming principals, principals at all three types of schools have roughly the same amount of average teaching experience, about 14 years. Principals at any type of school are, on average, 49 to 50 years old.

7 Results are based on multivariate regression analyses.


3. How Do Teachers in High-LEP Schools Differ from Teachers in Low- and No-LEP Schools?

Together with principal leadership, teacher quality has been cited as the most important factor in determining school effectiveness and, ultimately, student achievement (Clewell and Campbell 2004; Jordan, Mendro, and Weerasinghe 1997; Sanders and Rivers 1996; Wright, Horn, and Sanders 1997). This section describes the characteristics and backgrounds of teachers in the three types of schools under study to explore potential differences in preparation, certification, experience, and professional development.

Teacher Demographics and Compensation

The teaching force in High-LEP schools is more diverse than that of Low- or No-LEP schools, in both gender and race (table 7). Teachers in High-LEP schools appear not to differ from teachers in Low- or No-LEP schools in terms of age (mean of approximately 42 years for all three groups). High-LEP school teachers are, however, slightly more likely to be male (15 percent versus 11 percent), substantially more likely to be Asian (7 percent versus less than 2 percent), and more than five times more likely to be Hispanic (25 percent versus 2 to 5 percent) than those in schools with fewer LEP students. Overall, 40 percent of Asian teachers and 45 percent of Hispanic teachers nationwide teach in High-LEP schools. These patterns remain largely the same among new teachers, with one notable exception. The disproportionate concentration of male and minority teachers in High-LEP schools slightly increases among new teachers when comparing them across LEP school types.

Teachers in High-LEP schools earn more, on average, than do those in Low- or No-LEP schools.8 This finding is true regardless of amount of teaching experience, educational level (degrees earned), certification status, and school location. Although these experiential and demographic factors are the main predictors of teacher salaries, analysis reveals that a school’s LEP status also results in salary premiums. The average premium for teachers in High-LEP schools is over $2,000 compared with the earnings of teachers at Low-LEP schools and over $3,000 compared with teachers in schools with no LEP children.

Teacher Experience, Education, and Certification

High-LEP schools have more new teachers than schools with fewer or no LEP students (table 8). New teachers—those teaching for fewer than three years—represent 21 percent of teachers in High-LEP schools, 16 percent of those in Low-LEP schools, and 14 percent of those in No-LEP schools. In light of these differences, it is not surprising that High-LEP school teachers have, on average, slightly less teaching experience and time at their current schools than do teachers at Low- and No-LEP schools. The difference amounts to approximately one year between High- and Low-LEP schools and two years between High- and No-LEP schools, in time at current school and overall years of teaching experience.9

There is no difference in teacher mobility and attrition among schools with different concentrations of LEP students. But new teachers are more likely to change schools than the national teaching population (table 8). Patterns of teacher mobility and attrition are relatively

8 Based on multivariate regression analyses. 9 Researchers seem agree that years of teaching experience contribute to teacher quality during the first five years of teaching (Darling-Hammond 2000; Goldhaber and Anthony 2003).


similar among teachers in each of the three types of schools, whether examining all teachers or new teachers only. The data, which measure staffing changes after a one-year interval, suggest that approximately 84 percent of all teachers remain at the same school one year later, while about 8 percent switch schools and nearly 8 percent leave the profession. New teachers, however, change schools at greater rates than the general teaching population. On average, about 15 percent of new teachers change schools, versus about 8 percent of the general teaching force (including new teachers).

Overall, teachers in High-LEP schools have less academic preparation than their counterparts in other schools, but no such differences are found among new teachers at High- and No-LEP schools (table 8). Sixty-six percent of teachers in High-LEP schools indicate their highest degree is a bachelor’s degree, versus about 55 percent of those in other schools. In addition, only 33 percent of teachers in High-LEP schools hold a master’s degree, while about 45 percent of teachers in Low- and No-LEP schools do. Among new teachers—those who entered teaching within three years of being surveyed—differences are less marked. New teachers at No- and High-LEP schools are equally likely to hold a BA (85 percent do) or an MA (13–14 percent do), while new teachers at Low-LEP schools are more likely to hold an MA (20 percent) and therefore less likely to hold a BA (80 percent).10

ESL/bilingual education teachers in High- and Low-LEP schools are equally likely to be certified in that field, but teachers in High-LEP schools are more likely to hold ESL/bilingual certification in addition to their main certification (table 10). When defining “certification” as having a regular certificate in a given field (excluding provisional, emergency, and temporary certification), more than 90 percent of ESL/bilingual teachers in both High- and Low-LEP schools are certified in that field. Teachers whose secondary assignments are ESL/bilingual also are certified at similar rates across school types, although in their case only about 60 to 70 percent are certified. Teachers at High-LEP schools are, however, more likely to hold an ESL/bilingual certificate even though this is neither their first nor the second teaching assignment (about 8 percent of High-LEP versus 2 percent of Low-LEP teachers). As a result, when comparing all teachers at High-versus Low-LEP schools, a larger percentage of High-LEP (15 percent) than Low-LEP (4 percent) teachers is certified in ESL/bilingual education.

Teachers in High-LEP schools are more likely to have provisional, emergency, or temporary certification than are those in other schools (table 8). Teachers in High-LEP schools are not markedly less likely to hold certification in their main teaching field than those in schools with fewer LEP students. Large differences emerge, however, when the type of certification is examined. Teachers in High-LEP schools are substantially less likely to have full or probationary certification (indicating that they have completed a traditional teacher preparation program) and are more likely to have provisional, temporary, or emergency certification. This is partly because a greater proportion of teachers in High-LEP schools is new to the profession.

New teachers in High-LEP schools are substantially more likely to be uncertified than those at Low- or No-LEP schools (table 8, figure 4). When examining new teachers only, certification differences reported earlier among all teachers become even more pronounced. Overall, about 80 percent of new teachers in High-LEP schools are certified in their main field, versus about 90 percent of teachers in the other schools. This 10 percent difference becomes much larger, however, when type of certification is taken into account. Slightly more than 50 percent of teachers in High-LEP schools have full certification, while almost 80 percent of teachers at Low- and No-LEP schools do. In addition, new teachers at High-LEP schools are three times more likely to be uncertified (but

10 Research on the effect of degree level on teacher quality has produced mixed findings, and it is not clear whether an advanced degree contributes to teacher quality (Hanushek 1986; Greenwald, Hedges and Laine 1996).


in a certification program) and are twice as likely to have provisional, temporary, or emergency certification than new teachers at Low- or No-LEP schools. High-LEP schools are, therefore, at a double disadvantage: they have a larger share of new teachers and their new teachers are less likely to be certified.11

Figure 4. Certification Status of New Teachers

Source: Urban Institute analysis of NCES Schools and Staffing Survey, 1999-2000

21% New Teachers

14% New Teachers28% Provisional Certification

19% Uncertified

7% Uncertified

13% Provisional Certification

80% Certified76% Certified

High LEP

16% New Teachers

No LEPLow LEP

53% Certified

15% Provisional Certification

9% Uncertified

Teacher In-Service Professional Development

Teachers in High-LEP schools tend to report receiving more professional development than do teachers in other types of schools (tables 9 and 10, figure 5). In three of four professional development (PD) activities where differences are found, higher proportions of teachers in High-LEP schools report participation than teachers in Low- and No-LEP schools. The subject of this PD training also differs slightly by whether the schools have high concentrations of LEP students. Teachers in High-LEP schools are more likely to have participated in PD in the subject matter of their main teaching field in the past year, teaching methods/pedagogy, and student assessment. Conversely, they are substantially less likely to have had PD in the use of computers for instruction than are teachers in the other types of schools. Lastly, much higher proportions of general education teachers (those who do not specifically teach bilingual education or ESL) received training geared toward teaching LEP students in High-LEP schools (63 percent) than is true in No- (15 percent) or Low-LEP schools (25 percent).

11 Findings of research on the effect of licensure on teacher quality are mixed, with some studies judging fully licensed teachers more effective (Evertson, Hawley, and Zlotnik 1985) and others finding little effect (Goldhaber and Brewer 2000; Walsh 2001).


Figure 5. In-service Professional Development of Teachers

High Low High Low High Low High Low High Low High Low

Source: Urban Institute analysis of NCES Schools and Staffing Survey, 1999-2000. See Table 9 in Appendix B. Note: * no significant difference between High and Low LEP schools.

Teachers in Schools by LEP Concentration

Content Area (main teaching

field)

Management (of classrooms)*

42% 42%

79%

Methods (of teaching)

Assessment (of students)

Computers (use in

instruction)

71%

63%

83%

71%

Standards (main teaching

field)*

77%73%

68%

83% 80%


CONCLUSIONS

This report compares schools with varying concentrations of limited English proficient (LEP) children, focusing on three critical components: the characteristics of the schools themselves, their principals, and their teachers. The following summarizes our key findings for each unit of analysis:

School Characteristics

• The most salient finding is that LEP students are highly concentrated in a few schools.12 Nearly 70 percent of LEP students are enrolled in 10 percent of elementary schools. In these schools, classified in this study as High-LEP schools, nearly 50 percent of the students are LEP.

• High-LEP schools are more likely to be located in urban areas and therefore have many characteristics associated with urban schools: larger enrollments; larger class sizes; greater racial and ethnic diversity; higher incidences of student poverty, student health problems, tardiness, absenteeism, and lack of preparation; greater difficulty filling teaching vacancies; greater reliance on unqualified teachers; and lower levels of parent involvement. As a consequence of higher poverty levels, High-LEP schools are more likely to offer Title I services than their Low-LEP and No-LEP counterparts.

• Perhaps because of the greater concentration of at-risk students and their eligibility for Title I funding, High-LEP schools are more likely to offer support and remedial programs, such as pre-K, enrichment, after-school, and summer school programs. They are also more apt to be involved in parental outreach and support activities.

• Likely due to high enrollment of LEP students, High-LEP schools are also more likely to use standardized processes for identifying LEP students and to provide specialized instruction for non-English speakers, such as foreign language immersion programs, bilingual education, ESL, and native language instruction. High-LEP schools are three times more likely to use native language instruction than Low-LEP schools.

Principal Characteristics

• On average, principals in High-LEP schools have less education and training than those in other schools. They are more likely to hold an MA or a PhD, but less likely to have a post-MA (non-PhD) education specialist degree, than principals in Low- and No-LEP schools. They have served as principals in their current schools for less time than their counterparts at other schools, and report fewer years of experience as principals than No-LEP school principals. High-LEP principals do, however, have roughly the same amount of teaching experience as do principals in both other types of schools.

• Principals in High-LEP schools are more likely to be Hispanic, Asian, and female than principals in Low-LEP or No-LEP schools. High-LEP school principals are roughly the same age as their counterparts in both other types of schools. On average, however, they earn higher salaries than those in Low-LEP and No-LEP schools.

12 This high concentration of LEP students is mentioned in studies by Ruiz-de-Velasco and Fix (2000) and Orfield (2001).


Teacher Characteristics

• Teachers in High-LEP schools report having less academic preparation than their Low-LEP and No-LEP counterparts (although there is little difference among new teachers); they are more likely to have provisional, emergency, or temporary certification. High-LEP schools have a greater proportion of new teachers than do schools with fewer LEP students, and these teachers are much more likely to be uncertified.

• Teachers in High-LEP schools report receiving more professional development in content, methods and assessments than do teachers in the other two types of schools. Furthermore, non-ESL/bilingual teachers in High-LEP schools were more likely to report receiving at least eight hours of LEP-focused professional development in the past three years than were their counterparts in Low-LEP schools.13

• The teaching force in High-LEP schools tends to be more racially and ethnically diverse than that of Low-LEP or No-LEP schools and is more likely to be male. On average, teachers in High-LEP schools earn more than their counterparts at other schools.

DISCUSSION

The findings of this study offer a highly complicated picture of the schools where most LEP and immigrant students are educated. Adding to this complexity are the interactions among the effects of poverty, language minority status, ethnicity, and urbanicity that converge as we attempt to describe the educational landscape for LEP students. How much of the difference between High-LEP schools and their Low-LEP and No-LEP counterparts can be attributed to the fact that High-LEP schools are also poor, urban schools and how much can be attributed to the concentrated presence of LEP students enrolled in these schools?

It is probably safe to say that many differences attributable to the presence of LEP students are those where schools have chosen to respond to the challenges and conditions of their urban settings by specifically addressing the needs of LEP students. For example, the high incidence of schoolwide Title I funding in urban schools means High-LEP schools are likely to have funds for special programs, such as parent outreach activities.14 Many of them may have chosen to use these funds to involve parents of LEP students by providing interpreters and translations of letters that go to LEP parents. Also, the presence of a critical mass of LEP students, owing to the extreme concentration of these students in relatively few schools, has probably contributed to the use of approaches and strategies considered effective for this population: standardized identification procedures, support programs (remedial and enrichment), and specialized instruction for ELLs (such as bilingual education, ESL, foreign language immersion programs, and native language instruction).

Greater racial and ethnic diversity among High-LEP teachers can be an effect both of urbanicity and of the fact that ESL/bilingual teachers are more likely to be Hispanic or Asian.15 Some teachers’ lack of professional credentials in High-LEP schools can be attributed to the difficulty of filling positions in urban schools and to the shortage of ESL/bilingual teachers nationally.16 The lack of credentials

13 This finding echoes a study based on 1993–94 data (McCandless, Rossi, and Daugherty 1996). 14 In fact, schools receiving Title I funding are required to spend a portion of the funds on parental outreach. 15 Thirty-eight percent of ESL/bilingual teachers are Hispanic and 4 percent are Asian, compared with 5 and 2 percent, respectively, of non-ESL/bilingual teachers. 16 This is supported by the greater number of new teachers in High-LEP schools. These teachers, who tend to be not fully certified in their main field of teaching, may have been hired to f ill acute shortages.


may, in turn, explain the higher rate of professional development provided to teachers in High-LEP schools. The shortage of bilingual/ESL teachers and their placement in urban schools can also help explain higher salary levels for new teachers in High-LEP schools. Schools, especially urban schools, have been known to offer monetary incentives to attract teachers, particularly in shortage areas.

The extremely high concentration of LEP students in urban schools forces us to define their education within an urban context. When examining school composition and policies, it is difficult to separate the effects of urbanicity and its attendant demographics—poverty, racial and ethnic diversity, teacher shortages, large enrollments—from the effects of LEP students and their needs. Our findings regarding how High-LEP schools differ from Low-LEP and No-LEP schools must therefore be viewed through the lens of what is known about urban schools. It is a two-way street: LEP students’ special needs may exacerbate educational challenges in urban schools, while the conditions present in urban schools may complicate the educational opportunities of LEP students.

IMPLICATIONS

What are the implications of our findings for the education of young immigrant students? The fact that LEP students are highly segregated has both positive and negative implications for these students’ educational opportunities. On the positive side, the density of LEP enrollment makes the provision of specialized services more cost-effective and a higher priority, which enhances the likelihood that such services will be offered. It is often easier to justify expenditures for special programs when a large proportion of the student body will benefit. On the negative side, the segregation of LEP students results in their isolation from the educational mainstream and the attendant loss of the benefits of interacting with English-speaking classmates. This lack of interaction also marks a loss for English-dominant students.

In terms of resources, the shortage of teachers in High-LEP schools with experience, adequate academic preparation, and appropriate credentials poses the most significant problem for LEP students. Efforts to recruit, prepare, and credential more ESL/bilingual teachers through colleges of teacher education and alternative certification programs can help address this shortage. Additionally, research has pointed to the need to improve teacher education programs for ESL/bilingual teachers and to provide all non-ESL/bilingual teachers with training in working with LEP students (GAO 1994; Menken and Antunez 2001). This training is particularly needed at Low-LEP schools, where this analysis shows teachers are less likely to receive in-service professional development related to teaching LEP students.

Finally, this research has identified a particularly serious area of concern: the lower availability of LEP-geared services at Low-LEP schools, where a large share of LEP students enroll. Low-LEP schools are less likely to offer special support and enrichment programs than other schools. In addition, instruction adapted to the needs of LEP children is less prevalent at these schools where, as mentioned above, general education (non-ESL/bilingual) teachers are also less likely to receive in-service professional development related to teaching LEP students. Nearly one-third of LEP students attend these mostly suburban schools and are likely taught by these teachers. Moreover, within individual schools, LEP students are relatively isolated. The expansion of the immigrant population to nontraditional, secondary sites is causing LEP student enrollment to become more diffuse across schools, further increasing the extent of isolation within schools.17 As this process of

17 As mentioned earlier, in recent years, immigrants have tended to settle not in traditional receiving states (such as California and Florida) but in other parts of the country (such as Colorado and Mississippi). Recent research suggests, however, that the trend toward segregation of LEP students may be continuing in the so-called “new growth” states where many immigrants moved in the 1990s (Fix and Passel 2003).


immigrant expansion throughout the nation continues to unfold, it will be of particular importance for all schools to offer the services needed to help LEP children succeed. Perhaps a greater emphasis on exposing non-ESL/bilingual teachers to the needs of LEP students through both pre-service and in-service training can be a starting point to address this problem, and will help them become more effective teachers of LEP students.

In the specific context of NCLB, what are the implications of our findings for LEP students? Our findings are based on data collected before the passage of the No Child Left Behind Act and thus provide good baseline data against which to measure the potential effect of this legislation on High-LEP and Low-LEP schools. The increased accountability requirements of NCLB relating to the LEP subpopulation of students may already be changing the educational landscape for these students. To explore that specific question, the Urban Institute is conducting a series of case studies of High-LEP schools and school districts to assess the effect of NCLB on the instruction of LEP students. From the findings of this analysis of SASS data, however, it is possible to speculate what some of these effects may be:

Increased resources for LEP students in High-LEP schools and potentially decreased resources for LEP students in Low-LEP schools. Because of the heightened accountability requirements for subpopulations of students and the increased difficulty of exempting students from testing requirements, High-LEP schools may offer increased attention, resources, and special services for LEP students.18 This may also be true in Low-LEP schools that have moderate LEP populations. At the same time, in schools where LEP students are isolated (in numbers too small to require that test scores be reported separately under NCLB), these students run the risk of being overlooked while resources are concentrated on other subgroups of at-risk students.

Multiple NCLB reporting categories affecting immigrant students—language, poverty, and race or ethnicity.19 Children who are limited English proficient are also likely to be racial or ethnic minorities and economically disadvantaged. As such, these students may come under NCLB’s accountability requirements multiple times. LEP students in Low-LEP schools who might be overlooked in the LEP category because of low representation may receive attention because of their ethnicity or poverty status. On the other hand, schools that serve LEP students who fall into multiple NCLB categories may encounter difficulties addressing the multiple disadvantages posed by different conditions (say, language and poverty) that may require different types of interventions.

Worsening of the already-critical shortage of bilingual/ESL teachers. In the short term, NCLB requirements for “highly qualified” teachers and paraprofessionals may exacerbate the current shortage of bilingual and ESL teachers and paraprofessionals. In the long run, however, the law might improve the quality of teachers that serve LEP students.

Increased attention and resources for parental involvement. NCLB requires that schools encourage parent involvement in choices regarding students’ programs for English language acquisition.20 As a result, schools may initiate or expand activities and programs for parents of LEP students and adopt practices (such as translating letters into the dominant foreign language) to foster outreach.

18 Increased federal funding under Title III for providing language instruction to LEP students may contribute to this increase in resources. 19 Disability might also be another category, but one excluded from the present discussion because it is outside the scope of this paper. 20 See the Department of Education’s “Declaration of Rights for Parents of English Language Learners,” http://www.ncela.gwu.edu/resfor/parents/parent_bill.pdf


REFERENCES

Bryk, Anthony, Penny Bender Sebring, David Kerbow, Sharon Rollow, and John Easton. 1998. Charting Chicago School Reform. Boulder, CO: Westview Press.

Capps, Randy, Michael Fix, Julie Murray, Jason Ost, Jeffrey Passel, and Shinta Herwantoro. 2005. The New Demography of America’s Schools: Immigration and the No Child Left Behind Act. Washington, DC: The Urban Institute.

Clewell, Beatriz, and Patricia Campbell. 2004. Highly Effective USI Schools: An Outlier Study. Washington, DC: The Urban Institute/Campbell-Kibler Associates, Inc.

Darling-Hammond, Linda. 2000. “Teacher Quality and Student Achievement: A Review of State Policy Evidence.” Education Policy Analysis Archives 8(1).

Evertson, Carolyn, Willis Hawley, and Marilyn Zlotnik. 1985. “Making a Difference in Educational Quality through Teacher Education.” Journal of Teacher Education 36(3): 2–12.

Fix, Michael, and Jeffrey Passel. 2003. “U.S. Immigration: Trends and Implications for Schools.” Washington, DC: The Urban Institute. http://www.urban.org/url.cfm?ID=410654.

General Accounting Office (GAO). 1994. Limited English Proficiency: A Growing and Costly Educational Challenge Facing Many School Districts. GAO/HEHS-94-38. Washington, DC: GAO.

Goldhaber, Dan, and Emily Anthony. 2003. “Teacher Quality and Student Achievement.” Urban Diversity Series No. 115. ERIC Clearinghouse on Urban Education. New York: Teachers College, Columbia University.

Goldhaber, Dan, and Dominic Brewer. 2000. “Does Teacher Certification matter? High-School Teacher Certification Status and Student Achievement.” Educational Evaluation and Policy Analysis 22(2): 129–45.

Greenwald, Rob, Larry Hedges, and Richard Laine. 1996. “The Effect of School Resources on Student Achievement.” Review of Educational Research 66(3): 361–96.

Hallinger, Philip, and Ronald Heck. 1996. “Reassessing the Principal’s Role in School Effectiveness: A Review of Empirical Research.” Educational Administration Quarterly 32(1): 5–44.

———. 1998. “Exploring the Principal’s Contribution to School Effectiveness: 1980–1995.” School Effectiveness and Improvement 9:157–91.

Hanushek, Eric. 1986. “The Economics of Schooling: Production and Efficiency in Public Schools.” Journal of Economic Literature 24(3): 1141–78.

Jordan, H.R., R.L. Mendro, and D. Weerasinghe. 1997. “Teacher Effects on Longitudinal Student Achievement.” Paper presented at the National Evaluation Institute, Indianapolis, July.

McCandless, Elise, Robert Rossi, and Shannon Daugherty. 1996. “Are Limited English Proficient Students Being Taught by Teachers with LEP Training?” Issue Brief 7-96. Washington, DC: National Center for Education Statistics.


Menken, Kate, and Beth Antunez. 2001. An Overview of the Preparation and Certification of Teachers Working with Limited English Proficient (LEP) Students. ERIC # ED455231. Washington, DC: National Clearinghouse for Bilingual Education.

Orfield, Gary. 2001. Schools More Separate: Consequences of a Decade of Resegregation. ERIC # ED459217. Cambridge: Harvard Civil Rights Project.

Padolsky, Daniel. 2005. “How Many School-Aged English Language Learners (ELLs) Are There in the U.S.?” Ask NCELA No. 1. Revised May 2005. Washington, DC: National Clearinghouse for English Language Acquisition and Language Instruction Educational Programs. http://www.ncela.gwu.edu/expert/faq/01leps.htm.

Ruiz-de-Velasco, Jorge, and Michael Fix, with Beatriz Chu Clewell. 2001. Overlooked and Underserved: Immigrant Students in U.S. Secondary Schools. Washington, DC: The Urban Institute.

Sanders, William, and June Rivers. 1996. Cumulative and Residual Effects of Teachers on Future Student Academic Achievement. Research Progress Report. Knoxville: University of Tennessee Value-Added Research and Assessment Center.

Sebring, Penny Bender, and Anthony Bryk. 2000. “School Leadership and the Bottom Line in Chicago.” [Electronic version]. Phi Delta Kappan 81(6): 440–43.

Walsh, Kate 2001. Teacher Certification Reconsidered: Stumbling for Quality. Baltimore, MD: Abell Foundation.

Waters, Tim, Robert Marzano, and Brian McNulty. 2003. Balanced Leadership: What 30 years of research tells us about the effect of leadership on student achievement. A Working Paper. Aurora, CO: McREL.

Wright, S. Paul, Sandra Horn, and William Sanders. 1997. “Teacher and Classroom Context Effects on Student Achievement: Implications for Teacher Evaluation.” Journal of Personnel Evaluation in Education 11(1): 57–67.

____________________________________________________________________________________ THE URBAN INSTITUTE Program for Evaluation and Equity Research (PEER)

APPENDIX A:

LIST OF FIGURES AND TABLES


FIGURES Figure 1: Concentration of LEP Students in Elementary Schools Figure 2: Demographic Description of Elementary Schools Figure 3: Special Programs Offered Figure 4: Certification Status of New Teachers Figure 5: In-Service Professional Development of Teachers TABLES Table 1: Elementary Schools by LEP Enrollment Table 2: Characteristics of Public Elementary Schools Table 3: Instruction in Public Elementary Schools Table 4: Problems Reported by Staff in Public Elementary Schools Table 5: Parent Participation and Teaching Vacancies in Public Elementary Schools Table 6: Principal Characteristics in Public Elementary Schools Table 7: Demographics of Teachers in Public Elementary Schools Table 8: Teaching Experiences, Training, and Certification of Teachers in Public Elementary

Schools Table 9: Professional Development of Teachers in Public Elementary Schools Table 10: Teacher Certification, Teaching Assignment, and Professional Development


APPENDIX B:

TABLES

N % N % N %

All 49,826 100 22,637,370 100 2,049,678 100

LEP CategoryNo-LEP 21,277 42.7 7,596,360 33.6 0 0Low-LEP 23,566 47.3 12,040,539 53.2 643,492 31.4High-LEP 4,983 10.0 3,000,471 13.3 1,406,186 68.6


Notes: No-LEP = no LEP children enrolled at these schools; low-LEP = < 23.48% children are LEP; high-LEP = =23.48% of children are LEP.

LEP Enrollment

Table 1: Elementary Schools by LEP Enrollment

Elementary Schools Total Enrollment

No-LEP Low-LEP High-LEP All schools21,277 23,566 4,983 49,82642.7 47.3 10.0

0.00 5.04 46.88 7.070 1.45 1.44

357.03 510.92 602.16 454.3323.67 23.98 22.73

20.78 27.15 46.1 26.3238.36 54.34 43.89 46.4740.87 18.51 10.02 27.21

12.5 14.5 39.28 16.422.51 2.48 2.36

14.81 16.15 17.86 15.750.32 0.32 0.29

75.65 65.17 22.98 65.432.19 2.1 1.8924.35 34.83 77.02 34.572.19 2.1 1.89

0.88 3.87 9.93 3.21.53 1.55 1.5218.68 15.64 10.86 16.461.55 1.35 1.193.25 14.05 52.75 13.32.39 2.44 2.331.55 1.27 3.48 1.610.38 0.36 0.35

41.34 39.86 71.87 43.692.28 2.23 2.07

30.01 40.13 18.30 33.6219.98 16.33 15.59 17.8150.01 43.54 66.18 48.56

148,561 196,801 92,715 438,07721.08 19.8 34.86 21.73

-- 29.03 53.99 18.05

24.89 25.02 51.51 26.4136.88 37.79 60.38 39.3452.98 54.31 80.22 57.25

Source: Urban Institute analysis of NCES Schools and Staffing Survey 1999-2000 (school and teacher surveys).

Notes: Standard deviations calculated using Taylor Series Linearization.

Difference from high-LEP: p < .0001 p = .01 p = .05

Title I Services

Type of Title I Services Offered at SchoolNoneTitle I targeted assistance

by Type of Title I Services OfferedNo Title I services

Title I schoolwide

Targeted Assistance Schools: % Students ServedN served% served

Title I targeted assistanceTitle I schoolwide

Schools%

Mean % LEP

School Characteristics and Student Demographics

SD

Students receive Title I services in ESL

Percent Free/Reduced-Price Lunch Eligible

Mean EnrollmentSD

Urbanicity of SchoolUrbanSuburbanRural

Mean Percent Minority Teachers in SchoolSD

Mean Students per TeacherSD

Student Racial ProfileWhite

BlackSD

SDAll Minorities

SD

Minority Distribution (as percent of LEP category)

Mean % Free/Reduced-Price Lunch EligibleSD

Table 2: Characteristics of Public Elementary Schools

HispanicSD

Native AmericanSD

AsianSD

No-LEP Low-LEP High-LEP All schools

30.31 33.06 42.86 32.88

4.83 7.15 9.25 6.3759.08 57.28 65.55 59.0756.43 58.80 71.19 59.84

68.02 72.66 69.51 70.37

5.20 11.60 27.17 10.42

49.36 59.38 73.06 56.47

65.03 73.80 82.46 70.92

22.40 25.33 34.20 24.96

38.97 37.73 34.79 37.99

93.83 88.78 92.9486.03 75.88 84.2677.49 94.78 80.5172.74 67.82 71.8887.99 86.02 87.6539.54 51.36 41.677.92 95.52 80.99

85.23 98.81 87.6

81.66 94.07 83.8214.29 42.97 19.2977.42 89.89 79.59

11.89 43.50 17.4170.20 92.43 74.0780.15 83.15 80.67

Source: Urban Institute analysis of NCES Schools and Staffing Survey 1999-2000 (school and teacher surveys).

Notes : Standard deviations calculated using Taylor Series Linearization.


Information from parents

Table 3: Instruction in Public Elementary Schools

Pre-K on Site

Magnet program

Racial/Ethnic diversity is a goal

Special Programs

All students participate

Gifted and Talented program

Foreign language immersion

Before-/after-school academic enrichment

Summer school, academic intersessions (remedial)

Summer school, academic intersessions (for advancement)

Medical health services offered on site

LEP/ELL Policies and Procedures

How do you decide which students are LEP?

Teacher observation or referralHome language surveyStudent interviewStudent recordsAchievement testsLanguage proficiency tests

Yes (as a percent of all schools)

Does this school offer special instruction to meet the needs of LEP students?

Language Instruction of ELLsESL, structured immersion, or bilingual educationInstruction to maintain native languageInstruction in regular language arts classrooms

Subject Matter Instruction of ELLsIn native languageUsing ESL, bilingual, or immersion techniquesIn regular English-speaking classrooms

Table 4: Problems Reported by Staff in Public Elementary Schools

No-LEP Low-LEP High-LEP All schools No-LEP Low-LEP High-LEP All schools

Student TardinessSerious problem 3.23 3.88 7.19 3.93 5.73 6.75 10.48 6.78Moderate problem 18.89 24.14 32.19 22.71 20.91 26.69 33.89 25.36Minor problem 52.44 55.03 50.27 53.45 47.20 46.82 40.60 46.29Not a problem 25.45 16.96 10.35 19.90 26.16 19.74 15.03 21.58

Student AbsenteeismSerious problem 2.57 3.05 6.64 3.20 6.52 6.93 13.14 7.44Moderate problem 16.05 17.61 25.77 17.76 23.27 28.15 37.40 27.36Minor problem 53.06 56.32 55.25 54.83 46.05 46.18 38.34 45.29Not a problem 28.33 23.01 12.34 24.21 24.16 18.75 11.12 19.90

Unprepared StudentsSerious problem 12.68 12.78 29.98 14.44 23.08 23.99 39.30 25.30Moderate problem 36.21 34.58 34.17 35.23 32.52 30.81 36.08 32.00Minor problem 41.10 42.93 28.73 40.74 33.79 33.38 20.38 32.14Not a problem 10.02 9.71 7.13 9.58 10.62 11.82 4.24 10.57

PovertySerious problem 17.07 17.25 46.17 20.05 20.84 17.83 41.98 21.51Moderate problem 32.15 31.11 34.17 31.86 27.62 27.44 34.97 28.31Minor problem 35.38 35.41 16.51 33.52 33.17 34.88 15.83 32.22Not a problem 15.41 16.22 3.14 14.57 18.37 19.85 7.22 17.96

Student HealthSerious problem 2.73 2.16 6.81 2.86 4.49 3.82 12.55 5.00Moderate problem 14.10 15.01 26.15 15.73 16.73 16.99 32.30 18.54Minor problem 50.76 50.33 53.48 50.83 43.15 45.53 40.03 44.08Not a problem 32.41 32.50 13.56 30.58 35.63 33.66 15.11 32.39

Parental InvolvementSerious problem 8.35 9.00 14.79 9.30 19.38 19.44 30.34 20.58Moderate problem 28.51 28.15 41.13 29.59 30.00 29.59 37.30 30.56Minor problem 40.62 37.42 30.34 38.07 33.30 30.68 23.13 30.83Not a problem 22.51 25.44 13.75 23.03 17.32 20.29 9.23 18.02

Source: Urban Institute analysis of NCES Schools and Staffing Survey 1999-2000 (teacher and principal surveys).



Principals Teachers

Table 5: Parent Participation and Teaching Vacancies in Public Elementary SchoolsNo-LEP Low-LEP High-LEP All schools

Parent Participation

Does this school offer:-- 83.32 96.32 85.59-- 68.33 92.66 72.58-- 64.99 85.42 68.56

School Strategies for Increasing Parent Involvement:Does the school have:

A staff member assigned to work on parent involvement 54.9 60.28 75.73 59.53

A log of parent participation maintained by parents or staff 68.1 80.43 83.52 75.47

A reliable system of communication with parents, such as phone trees 97.92 98.01 95.89 97.76

40.73 50.8 65.65 47.99

A parent drop-in center or lounge 23.2 28.34 36.43 26.95

Requirement that teachers send letters explaining lessons to parents 56.24 59.17 67.44 58.75

Requirement that teachers suggest activities parents do at home w/children 50.12 54.71 62.61 53.54

Requirement that teachers create assignments that involve parents 32.95 36.35 53.53 36.62

Teaching Vacancies% of all schools with vacancies 75.87 86.19 82.21 81.38

Did the school use the following methods to fill vacancies? (of those w/ vacancies)

Hired a fully qualified teacher 93.96 95.83 81.42 93.63

Hired a less-than-fully qualified teacher 10.86 11.38 36.51 13.71

Canceled planned course offerings 0.64 1.54 2.79 1.31

Expanded class sizes 7.76 9.57 8.1 8.7

Added sections to other teachers' normal teaching loads 2.31 2.75 3.67 2.67

Assigned a teacher of another subject or grade 4.18 6.26 7.39 5.55

Assigned an administrator or counselor to teach 2.49 1.42 1.83 1.89

Used a long-term or short-term substitute 22.47 28.13 47.38 27.82

Vacancies by Subject Area General Elementary (% that reported at least one vacancy) 67.14 80.76 75.74 74.44Difficulty to fill (of those w/vacancies)

Easy to fill 69.84 67.55 46.92 66.33Somewhat difficult to fill 25.24 25.99 38.92 27.08Very difficult to fill 4.40 5.52 12.48 5.8Could not fill the vacancy 0.34 0.95 1.69 0.79

ESL (% that reported at least one vacancy) 1.88 20.68 36.88 14.27Difficulty to fill (of those w/vacancies)


Special Education (% that reported at least one vacancy) 35.34 48.68 40.31 42.15Difficulty to fill (of those w/vacancies)


Source: Urban Institute analysis of NCES Schools and Staffing Survey 1999-2000 (school survey).



Outreach for LEP parents

Services to support involvement, such as child care or transportation

Interpreters for LEP parentsTranslations for LEP parents

Table 6: Principal Characteristics in Public Elementary SchoolsNo-LEP Low-LEP High-LEP All schools

Number of schools with principal information 19,381 21,740 4,542 45,663% of schools 91.09 92.25 91.15 91.64

Principal CharacteristicsMean Age 49.36 49.60 49.87 49.52SD 0.68 0.67 0.63

Race by LEP School DistributionAmerican Indian 0.53 0.50 1.15 0.58Asian or Pacific Islander 0.04 0.63 3.86 0.70Black 12.33 11.22 12.93 11.86Hispanic 3.01 4.36 24.21 5.76White 84.10 83.29 57.85 81.10

Race (national distribution)American Indian 38.91 41.4 19.7 --Asian or Pacific Islander 2.23 42.74 55.02 --Black 44.12 45.04 10.85 --Hispanic 22.17 36.02 41.81 --White 44.01 48.9 7.09 --

GenderMale 51.32 42.7 36.87 45.78Female 48.68 57.3 63.13 54.22

Highest DegreeBachelor's degree 1.57 1.66 1.97 1.65Master's degree 53.35 52.75 61.94 53.92Education specialist or professional diploma 36.65 35.33 22.69 34.63Doctorate or professional degree 8.43 10.26 13.4 9.8

Mean Years as Principal of Current School 6.01 5.01 4.24 5.36SD 0.41 0.4 0.36

Mean Years Total Principal Experience 10.36 8.78 8.57 9.43SD 0.69 0.67 0.62

Mean Years Teaching Experience 14.38 14.17 13.86 14.23SD 0.69 0.68 0.64

Source: Urban Institute analysis of NCES Schools and Staffing Survey 1999-2000 (principal survey).



Table 7: Demographics of Teachers in Public Elementary Schools


Teachers 495,985 719,429 145,667 1,361,081 69,079 114,035 30,580 213,694% of all teachers 36.44 52.86 10.70 13.93 15.85 20.99 15.70

Teacher Demographics

Mean Age 42.48 42.12 41.67 42.20 30.54 30.11 30.11 30.25SD 0.60 0.60 0.56 0.78 0.73 0.63

GenderMale 11.12 11.17 15.4 11.61 12.8 14.56 18.15 14.51Female 88.88 88.83 84.6 88.39 87.2 85.44 81.85 85.49

Race by LEP School DistributionAmerican Indian 0.65 0.71 1.71 0.79 0.98 0.81 1.20 0.92Asian or Pacific Islander 0.39 1.88 6.89 1.87 1.09 1.44 8.03 2.27Black 9.35 6.91 7.59 7.87 11.67 7.12 10.05 9.01Hispanic 1.90 4.87 24.54 5.89 1.98 8.76 30.01 9.61White 87.71 85.64 59.27 83.57 84.28 81.87 50.71 78.19

Race (national distribution)American Indian 29.78 47.15 23.07 -- 34.50 46.87 18.63 --Asian or Pacific Islander 7.51 53.11 39.39 -- 15.51 33.89 50.60 --Black 43.28 46.39 10.32 -- 41.87 42.16 15.97 --Hispanic 11.77 43.65 44.57 -- 6.65 48.66 44.69 --White 38.25 54.16 7.59 -- 34.84 55.88 9.28 --

Source: Urban Institute analysis of NCES Schools and Staffing Survey 1999-2000 (teacher survey).



All Teachers New Teachers (fewer than 3 years)

Table 8: Teaching Experiences, Training, and Certification of Teachers in Public Elementary Schools


Teachers 495,985 719,429 145,667 1,361,081 69,079 114,035 30,580 213,694% 36.44 52.86 10.70 13.93 15.85 20.99 15.70

Teaching Experiences

Mean Years at Current School 9.85 8.59 7.44 8.93SD 0.38 0.36 0.33

Mean Years Teaching Experience 15.44 14.35 13.07 14.61SD 0.54 0.53 0.49

Teacher AttritionStayers 84.90 84.11 82.97 84.28 74.54 75.63 80.71 76.00Movers 7.89 8.32 8.27 8.16 15.91 15.28 9.98 14.73Leavers 7.19 7.48 8.77 7.51 9.55 8.87 9.31 9.15

Training and Certification

Highest DegreeAssociate's or less 0.21 0.17 0.15 0.17 0.84 0.50 0.02 0.55Bachelor's 54.17 55.50 66.37 56.18 84.65 79.82 86.67 82.36Master's 45.39 43.75 32.80 43.18 14.45 19.65 12.64 16.96PhD or professional 0.23 0.58 0.68 0.47 0.06 0.03 0.67 0.13

Certification in Main Field (percent yes) 96.38 95.90 93.36 95.81 91.23 92.69 81.38 90.60

Full or probationary 91.70 91.91 83.43 90.93 76.44 79.82 52.74 74.85Provisional/temporary/emergency 4.68 3.99 9.93 4.88 14.79 12.87 28.64 15.75Not certified--in certification program 1.21 1.41 3.74 1.59 4.90 4.38 15.09 6.08Not certified--not in certification program 2.41 2.68 2.90 2.61 3.87 2.93 3.53 3.32




All Teachers New Teachers (fewer than 3 years)

By definition, fewer than 3 years teaching.

Table 9: Professional Development of Teachers in Public Elementary Schools

School Type No-LEP Low-LEP High-LEP All schools No-LEP Low-LEP High-LEP All schools

Teachers 495,985 719,429 145,667 1,361,081 69,079 114,035 30,580 213,694% of all teachers 36.44 52.86 10.70 13.93 15.85 20.99 15.70

Professional Development

In the past 12 months, have you participated in the following types of professional development?

Content area of main teaching field 62.80 68.05 72.77 66.64 -- -- -- --

Standards for main teaching field (content/performance) 76.60 79.73 82.90 78.93 -- -- -- --

Methods of teaching 73.54 78.61 82.80 77.21 -- -- -- --

Use of computers for instruction 70.44 71.28 62.47 70.03 -- -- -- --

Student assessment 65.51 70.51 77.01 69.38 -- -- -- --

Classroom discipline and management 42.41 41.64 41.93 41.95 -- -- -- --

Other professional development areas 20.26 20.04 23.47 20.49 -- -- -- --

Teacher Preparation Program Components

Did Teacher Prep Program Include…Coursework in instructional material selection -- -- -- -- 89.59 86.56 87.51 87.69Coursework in learning theory -- -- -- -- 95.02 95.01 92.11 94.58Observation of classroom teaching -- -- -- -- 96.89 97.35 94.46 96.77Feedback on own teaching -- -- -- -- 98.00 97.47 94.84 97.25

Practice TeachingNone -- -- -- -- 5.61 5.47 18.34 7.464 weeks or less -- -- -- -- 1.58 1.13 4.73 1.825-9 weeks -- -- -- -- 13.11 12.37 9.24 12.1410 weeks or more -- -- -- -- 79.69 81.03 67.68 78.59

Mentor in First Year of Teaching -- -- -- -- 66.03 67.54 68.58 67.20In same subject -- -- -- -- 49.59 55.01 50.99 52.68




New Teachers (fewer than 3 years)All Teachers

Table 10: Teacher Assignment, Certification, and Professional Development in ESL/Bilingual EducationPanel A: Teaching Assignment and Certification

Non ESL/Bilingual ESL/Bilingual Non ESL/Bilingual ESL/Bilingual

What is your main teaching assignment at this school?N 705,144 14,285 135,885 9,782% 98.0 2.0 93.3 6.7

% certified 96.0 91.3 93.4 92.8

Are you assigned to teach classes in another field at this school?N 45,980 5,673 10,440 2,440% 6.4 0.8 7.2 1.7

% certified 53.7 71.3 48.1 63.1

Do you hold any additional teaching certificates?N 126,719 13,491 24,131 11,272% 17.6 1.9 16.6 7.7

Certified ESL/Bilingual Teachers -- 30,549 -- 21,884

% of school category -- 4.2 -- 15.0

Panel B: LEP-Focused Professional Development

Non ESL/Bilingual ESL/Bilingual Non ESL/Bilingual ESL/Bilingual

N Teachers 342,056 19,617 112,193 12,222% of school category 47.5 2.7 77.0 8.4

Yes 25.351, 2 86.00 63.082 85.91No 74.651, 2 14.00 36.922 14.09

SOURCE: Urban Institute analysis of 1999-2000 Schools and Staffing Survey (Teacher Survey)

Difference from high-LEP: p <.0001 p = .01 p = .05

Low LEP High LEPM

ain

Assig

nmen

tSe

cond

ary

Assig

nmen

tAd

ditio

nal

Certi

ficat

ions

Sum

mar

y

Low LEP High LEP

NOTES (1) Different from High-LEP, p <.0001 (2) Different from ESL/Bilingual, p<.0001 (3) Indicated they have LEP students in their classrooms.

Teac

hers

Tea

chin

g LE

Ps3

Primary/Secondary Teaching Assignment

8 hours of LEP-focused PD in past 3 years? (%)

Documents

Immigrant Children in High and Low LEP Schools · 1 Throughout this research, LEP will include both students who are limited English proficient (LEP) and students who do not know